Visit Scotland Analysis

Setup

Load Libraries

library(tidyverse)

── Attaching core tidyverse packages ─────────────────────────────────────────────────────────── tidyverse 2.0.0 ──
✔ dplyr     1.1.0     ✔ readr     2.1.4
✔ forcats   1.0.0     ✔ stringr   1.5.0
✔ ggplot2   3.4.1     ✔ tibble    3.2.1
✔ lubridate 1.9.2     ✔ tidyr     1.3.0
✔ purrr     1.0.1     ── Conflicts ───────────────────────────────────────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag()    masks stats::lag()
ℹ Use the ]8;;http://conflicted.r-lib.org/conflicted package]8;; to force all conflicts to become errors

library(here)

here() starts at /Users/stuartstenhouse/Codeclan/visit_scotland_codeclan_project

library(sf)

Linking to GEOS 3.11.0, GDAL 3.5.3, PROJ 9.1.0; sf_use_s2() is TRUE

library(infer)
library(ggridges)
library(modelr)
library(skimr)
library(ggfortify)

Run Cleaning Script, Read In Clean Data & Set Colour Scheme for Visualisations

source(here("scripts/cleaning_script.R"))

regional_domestic_tourism_individual <- read_csv(here("data/clean_data/regional_domestic_tourism_individual_clean.csv"))
regional_domestic_tourism_non_gb_clean <- read_csv(here("data/clean_data/regional_domestic_tourism_non_gb_clean.csv"))
international_visits <- read_csv(here("data/clean_data/international_visits_clean.csv"))
tourism_businesses <- read_csv(here("data/clean_data/tourism_businesses_clean.csv"))
dom_int_summary <- read_csv(here("data/clean_data/dom_int_summary_clean.csv"))

local_authority_geo <- st_read(dsn = "data/geo_data/", layer = "pub_las")

colour_scheme <- c("#540453", "#1d1d65", "#970061", "#b6cb2b", "#9fd6f3")

Annual Average Stats

Domestic

5 Figure Summary

dom_int_summary %>%
  filter(dom_int == "Domestic") %>% 
  skim()

── Data Summary ────────────────────────
                           Values    
Name                       Piped data
Number of rows             11        
Number of columns          4         
_______________________              
Column type frequency:               
  character                1         
  numeric                  3         
________________________             
Group variables            None      

dom_int_summary %>%
  filter(dom_int == "International") %>% 
  skim()

── Data Summary ────────────────────────
                           Values    
Name                       Piped data
Number of rows             11        
Number of columns          4         
_______________________              
Column type frequency:               
  character                1         
  numeric                  3         
________________________             
Group variables            None      

IQR

dom_int_summary %>%
  filter(dom_int == "Domestic") %>%  
  summarise(visits_iqr= IQR(visits),
            expenditure_iqr= IQR(spend))

dom_int_summary %>%
  filter(dom_int == "International") %>%  
  summarise(visits_iqr= IQR(visits),
            expenditure_iqr= IQR(spend))

Median

dom_int_summary %>%
  filter(dom_int == "Domestic") %>%  
  summarise(median_visits = median(visits),
            median_spend = median(spend))

NA

dom_int_summary %>%
  filter(dom_int == "International") %>%  
  summarise(median_visits = median(visits),
            median_spend = median(spend))

Trends Over Time

Over Time / Visits

dom_int_summary %>% 
  ggplot() +
  geom_line(aes(x = year, y = visits, colour = dom_int),
          size = 2) +
  geom_point(aes(x = year, y = visits, colour = dom_int), 
             size = 4) +
  ylim(0, 15000) +
  labs(x = "\n Year",
       y = "Visitors (Thousands) \n",
       title = "Annual Visits",
       subtitle = "Domestic & International Overnight Vistors",
       colour = "") +
  scale_x_continuous(breaks = c(2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019)) +
  scale_colour_manual(values = c("Domestic" = "#540453",
                                 "International" = "#9fd6f3")) +
  theme(plot.title = element_text(size = 15, face = "bold"),
        plot.subtitle = element_text(size = 10),
        panel.background = element_rect(fill = "white"),
        panel.grid = element_line(colour = "grey90", 
                                  linetype = "dashed"))

Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
Please use `linewidth` instead.

Over Time / Expenditure

dom_int_summary %>% 
  ggplot() +
  geom_line(aes(x = year, y = spend, colour = dom_int),
          size = 2) +
  geom_point(aes(x = year, y = spend, colour = dom_int), 
             size = 4) +
  ylim(0, 4000) +
  labs(x = "\n Year",
       y = "Expenditure (Million GBP) \n",
       title = "Annual Expenditure",
       subtitle = "Domestic & International Overnight Vistors",
       colour = "") +
  scale_x_continuous(breaks = c(2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019)) +
  scale_colour_manual(values = c("Domestic" = "#540453",
                                 "International" = "#9fd6f3")) +
  theme(plot.title = element_text(size = 15, face = "bold"),
        plot.subtitle = element_text(size = 10),
        panel.background = element_rect(fill = "white"),
        panel.grid = element_line(colour = "grey90", 
                                  linetype = "dashed"))

Domestic Tourism

Average Visits By Region

regional_data_joined %>% 
  group_by(local_authority) %>% 
  summarise(mean_visits = mean(visits)) %>% 
  arrange(desc(mean_visits))

Simple feature collection with 32 features and 2 fields
Geometry type: MULTIPOLYGON
Dimension:     XY
Bounding box:  xmin: 5512.998 ymin: 530250.8 xmax: 470332 ymax: 1220302
Projected CRS: OSGB36 / British National Grid

regional_data_joined %>% 
  group_by(code) %>% 
  summarise(mean_visits = mean(visits)) %>% 
  ggplot(aes(fill = mean_visits)) + 
  geom_sf(colour = "grey90", size = 0.1) +
  labs(
    x = "Longitude",
    y = "Latitude",
    title = "Average Annual Number of Domestic Vistors",
    subtitle = "By Region",
    fill = "Visits (Thousands)") +
  theme(panel.background = element_rect(fill = "white"),
        axis.text.x = element_blank(),
        axis.text.y = element_blank(),
        axis.ticks = element_blank(),
        rect = element_blank(),
        axis.title.x = element_blank(),
        axis.title.y = element_blank())

Expenditure by Region

regional_data_joined %>% 
  group_by(local_authority) %>% 
  summarise(mean_expenditure = mean(expenditure)) %>% 
  arrange(desc(mean_expenditure))

Simple feature collection with 32 features and 2 fields
Geometry type: MULTIPOLYGON
Dimension:     XY
Bounding box:  xmin: 5512.998 ymin: 530250.8 xmax: 470332 ymax: 1220302
Projected CRS: OSGB36 / British National Grid

regional_data_joined %>% 
  group_by(code) %>% 
  summarise(mean_exp = mean(expenditure)) %>% 
  ggplot(aes(fill = mean_exp)) + 
  geom_sf(colour = "grey90", size = 0.1) +
  labs(
    x = "Longitude",
    y = "Latitude",
    title = "Average Annual Expenditure of Domestic Vistors",
    subtitle = "By Region",
    fill = "Expenditure (Million GBP)") +
    theme(panel.background = element_rect(fill = "white"),
        axis.text.x = element_blank(),
        axis.text.y = element_blank(),
        axis.ticks = element_blank(),
        rect = element_blank(),
        axis.title.x = element_blank(),
        axis.title.y = element_blank())

Expenditure Per Visit

Mean Expenditure Per Visit Plot

regional_domestic_tourism_individual %>%
  filter(visits != 0,
         expenditure != 0) %>% 
  mutate(exp_per_visit = expenditure / visits, rm.na = TRUE) %>%
  group_by(local_authority) %>% 
  summarise(mean_exp_per_visit = mean(exp_per_visit)) %>% 
  arrange(desc(mean_exp_per_visit)) %>%
  ggplot() +
  geom_col(aes(x = reorder(local_authority, mean_exp_per_visit), y = mean_exp_per_visit,
               fill = local_authority %in% c("City of Edinburgh", "Glasgow City", "Highland")),
           show.legend = FALSE, colour = "white") +
  scale_fill_manual(values = c("FALSE" = "grey80",
                               "TRUE" = colour_scheme[1])) +
  coord_flip() +
  labs(
    x = "Region",
    y = "Mean Expenditure per Visit (Million GBP / Thousand Visits)",
    title = "Domestic Tourism Mean Expenditure per Visit") +
  theme(plot.title = element_text(size = 15, face = "bold"),
        plot.subtitle = element_text(size = 10),
        panel.background = element_rect(fill = "white"),
        panel.grid = element_line(colour = "grey90", linetype = "blank"))

Nights Stayed

Plot

regional_domestic_tourism_individual %>% 
  mutate(nights_per_visit = nights / visits) %>% 
  group_by(local_authority) %>% 
  summarise(mean_nights_per_visit = mean(nights_per_visit, na.rm = TRUE)) %>%
  ggplot() +
  geom_col(aes(x = reorder(local_authority, mean_nights_per_visit), y = mean_nights_per_visit,
               fill = local_authority %in% c("Shetland Islands", "Orkney Islands", "Na h-Eileanan Siar",
                                             "City of Edinburgh", "Glasgow City", "Highland")),
           show.legend = FALSE, colour = "white") +
  scale_fill_manual(values = c("FALSE" = "grey80",
                               "TRUE" = colour_scheme[1])) +
  coord_flip() +
  labs(
    x = "Region",
    y = "Avg. Nights per Visit",
    title = "Domestic Tourism Avg. Nights per Visit") +
  theme(plot.title = element_text(size = 15, face = "bold"),
        plot.subtitle = element_text(size = 10),
        panel.background = element_rect(fill = "white"),
        panel.grid = element_line(colour = "grey90", linetype = "blank"))

Map

regional_data_joined %>%
  mutate(nights_per_visit = nights / visits) %>% 
  group_by(local_authority) %>% 
  summarise(mean_nights_per_visit = mean(nights_per_visit, na.rm = TRUE)) %>%
  ggplot(aes(fill = mean_nights_per_visit)) + 
  geom_sf(colour = "grey90", size = 0.1) +
  labs(
    x = "Longitude",
    y = "Latitude",
    title = "Domestic Tourism Mean Nights per Visit",
    fill = "Mean Nights per Visit"
  ) +
  theme(panel.background = element_rect(fill = "white"),
        axis.text.x = element_blank(),
        axis.text.y = element_blank(),
        axis.ticks = element_blank(),
        rect = element_blank(),
        axis.title.x = element_blank(),
        axis.title.y = element_blank(),
        plot.title = element_text(size = 15, face = "bold"))

NA

Gross Value Added

tourism_businesses %>%
  filter(local_authority != "Scotland",
         year %in% c(2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019)) %>%
  group_by(local_authority) %>% 
  summarise(total_gva = sum(gva)) %>% 
  slice_max(total_gva, n = 5)

  
tourism_businesses %>%
  filter(local_authority %in% c("City of Edinburgh", "Glasgow City", "Aberdeen City",
                                "Highland", "Fife"),
         year %in% c(2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019)) %>%
  ggplot() +
  geom_boxplot(aes(x = gva, y = local_authority, fill = local_authority), 
               alpha = 0.8,
               show.legend = FALSE) +
  scale_fill_manual(values = colour_scheme) +
  scale_colour_manual(values = colour_scheme) +
  labs(
    x = "Gross Value Added (Million GBP)",
    y = "Local Authority",
    title = "Distribution of Annual Gross Value Added",
    subtitle = "Top 5 Regions: 2009-2019"
  ) +
  theme(plot.title = element_text(size = 15, face = "bold"),
        plot.subtitle = element_text(size = 10),
        panel.background = element_rect(fill = "white"),
        panel.grid = element_line(colour = "grey90", 
                                  linetype = "dashed"))

Scottish Based Tourism Compared To English Based Tourism

regional_data_joined_sco_eng %>% 
  filter(region_of_residence == "Scotland") %>%
  group_by(local_authority) %>%
  summarise(mean_visits = mean(visits)) %>% 
  ggplot(aes(fill = mean_visits)) + 
  geom_sf(colour = "grey90", size = 0.1) +
  labs(
    x = "Longitude",
    y = "Latitude",
    title = "Average Annual Number of Scottish Based Vistors",
    subtitle = "By Region",
    fill = "Visits (Thousands)") +
  theme(panel.background = element_rect(fill = "white"),
        axis.text.x = element_blank(),
        axis.text.y = element_blank(),
        axis.ticks = element_blank(),
        rect = element_blank(),
        axis.title.x = element_blank(),
        axis.title.y = element_blank())

regional_data_joined_sco_eng %>% 
  filter(region_of_residence == "England") %>% 
  group_by(local_authority) %>%
  summarise(mean_visits = mean(visits)) %>% 
  ggplot(aes(fill = mean_visits)) + 
  geom_sf(colour = "grey90", size = 0.5) +
  labs(
    x = "Longitude",
    y = "Latitude",
    title = "Average Annual Number of English Based Vistors",
    subtitle = "By Region",
    fill = "Visits (Thousands)") +
  theme(panel.background = element_rect(fill = "white"),
        axis.text.x = element_blank(),
        axis.text.y = element_blank(),
        axis.ticks = element_blank(),
        rect = element_blank(),
        axis.title.x = element_blank(),
        axis.title.y = element_blank())

Marketing Edinburgh to Domestic Tourists: Mini Case Study

Visits by Country of Residence

regional_domestic_tourism_non_gb_clean %>%
  filter(local_authority %in% c("City of Edinburgh")) %>% 
  ggplot(aes(x = visits, y = region_of_residence, fill = region_of_residence == "Scotland")) +
  geom_density_ridges(show.legend = FALSE, alpha = 0.5) +
  theme_ridges() +
  theme(axis.title.y = element_blank()) +
  scale_fill_manual(values = c("FALSE" = colour_scheme[2],
                               "TRUE" = colour_scheme[1])) +
  labs(
    title = "Distribution of Annual Visits by Country of Residence",
    subtitle = "Edinburgh",
    x = "Visits (Thousands)") +
  theme(plot.title = element_text(size = 15, face = "bold"),
        plot.subtitle = element_text(size = 13),
        panel.background = element_rect(fill = "white"),
        panel.grid = element_line(colour = "grey90", linetype = "dashed"),
        axis.text.y = element_text(face = "bold"))

Expenditure Per Visit by Country of Residence

regional_domestic_tourism_non_gb_clean %>%
  filter(local_authority %in% c("City of Edinburgh")) %>%
  mutate(exp_per_visit = expenditure / visits) %>% 
  ggplot(aes(x = exp_per_visit, y = region_of_residence, fill = region_of_residence == "Scotland")) +
  geom_density_ridges(show.legend = FALSE, alpha = 0.5) +
  theme_ridges() +
  theme(axis.title.y = element_blank()) +
  scale_fill_manual(values = c("FALSE" = colour_scheme[2],
                               "TRUE" = colour_scheme[1])) +
  labs(
    title = "Distribution of Expenditure per Visit by Country of Residence",
    subtitle = "Edinburgh",
    x = "Expenditure per Visit (Million GBP / Thousand Visits)") +
  theme(plot.title = element_text(size = 15, face = "bold"),
        plot.subtitle = element_text(size = 13),
        panel.background = element_rect(fill = "white"),
        panel.grid = element_line(colour = "grey90", linetype = "dashed"),
        axis.text.y = element_text(face = "bold"))

Nights by Country of Residence

regional_domestic_tourism_non_gb_clean %>%
  filter(local_authority %in% c("City of Edinburgh")) %>% 
  mutate(nights_per_visit = nights / visits) %>%
  ggplot(aes(x = nights_per_visit, y = region_of_residence, fill = region_of_residence == "Scotland")) +
  geom_density_ridges(show.legend = FALSE, alpha = 0.5) +
  theme_ridges() +
  theme(axis.title.y = element_blank()) +
  scale_fill_manual(values = c("FALSE" = colour_scheme[2],
                               "TRUE" = colour_scheme[1])) +
  labs(
    title = "Distribution of Nights by Country of Residence",
    subtitle = "Edinburgh",
    x = "Nights") +
  theme(plot.title = element_text(size = 15, face = "bold"),
        plot.subtitle = element_text(size = 13),
        panel.background = element_rect(fill = "white"),
        panel.grid = element_line(colour = "grey90", linetype = "dashed"),
        axis.text.y = element_text(face = "bold"))

Linear Model: Predict Expenditure by Visits

regional_domestic_tourism_non_gb_clean %>% 
  filter(local_authority == "City of Edinburgh",
         region_of_residence == "England") %>% 
  summarise(cor(expenditure, visits))

edinburgh_england <- regional_domestic_tourism_non_gb_clean %>% 
  filter(local_authority == "City of Edinburgh",
         region_of_residence == "England")

model_edinburgh <- lm(formula = expenditure ~ visits, data = edinburgh_england)
summary(model_edinburgh)

Call:
lm(formula = expenditure ~ visits, data = edinburgh_england)

Residuals:
    Min      1Q  Median      3Q     Max 
-24.193  -4.918   4.838   6.423  14.165 

Coefficients:
              Estimate Std. Error t value Pr(>|t|)    
(Intercept) -493.61749   62.18315  -7.938 9.58e-05 ***
visits         0.62423    0.03945  15.824 9.76e-07 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 12.34 on 7 degrees of freedom
Multiple R-squared:  0.9728,    Adjusted R-squared:  0.9689 
F-statistic: 250.4 on 1 and 7 DF,  p-value: 9.756e-07

autoplot(model_edinburgh)

• Residuals vs. Fitted (Tests Independence of Residuals): No strong evidence of a pattern.
• Normal Q-Q (Tests Normality of Residuals): Distribution of standardised residuals appears fairly normal.
• Scale-Location (Tests Constancy of Variation of Residuals): No strong evidence of funneling.

edinburgh_england %>%
  add_predictions(model_edinburgh) %>%
  add_residuals(model_edinburgh) %>% 
  ggplot(aes(x = visits, y = expenditure)) +
  geom_point() +
  geom_line(aes(y = pred), col = colour_scheme[3], size = 1) +
  labs(
    x = "Visits (Thousands)",
    y = "Expenditure (Million GBP)",
    title = "English Based Visitors To Edinburgh: Expenditure / Visits",
    subtitle = "Based On 3 Year Annual Averages Between 2009-2019")

Model: spend ~ visits

Interpretation: ~97% of the variation in expenditure can be explained by the number of visits.

Question: The largest increase in visitor numbers is from 2009-2011 Avg. to 2010-2012 Avg. ~6%. If this percentage increase in visits was replicated in based on the most recent set of figures resulting in ~1803 (Thousand) visitors - what would the model predict for expenditure?

Answer: If visits were increased as above, the model would predict expenditure of ~632 Million GBP (Low Estimate ~620, High Estimate ~644)

As calculated below.

Step 1. outcome = b0 + b1 * 1st Predictor

Step 2. expenditure = b0(intercept) + b1(coefficient) * 1st Predictor

Step 3. expenditure = b0(intercept) + b1(coefficient) * visits

Step 4. expenditure = -493.61749 + 0.62423 * 1803

Step 5. expenditure = ~632 (Low Estimate ~620, High Estimate ~644)

632 + c(-12.34, 12.34)

[1] 619.66 644.34

International Tourism

Total Visitors by Country of Residence

international_visits %>% 
  select(-c(quarter, age, sample)) %>%
  filter(year %in% c(2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019),
         purpose == "Holiday") %>% 
  group_by(country) %>% 
  summarise(total_visits = sum(visits)) %>%
  slice_max(total_visits, n = 20) %>% 
  ggplot(aes(x = reorder(country, total_visits), y = total_visits)) +
  geom_col(aes(x = reorder(country, total_visits), y = total_visits,
               fill = country %in% c("USA")),
           show.legend = FALSE, colour = "white") +
  scale_fill_manual(values = c("FALSE" = "grey60",
                               "TRUE" = colour_scheme[1])) +
  coord_flip() +
  ylim(0, 3200) +
  labs(
    x = "Country",
    y = "Total Visits (Thousands)",
    title = "International Tourism: Total Visits (2002-2009)",
    subtitle = "Top 20 (Holiday Only)") +
  theme(plot.title = element_text(size = 15, face = "bold"),
        plot.subtitle = element_text(size = 10),
        panel.background = element_rect(fill = "white"),
        panel.grid = element_line(colour = "grey90", linetype = "blank"))

Total Expenditure by Country of Residence

international_visits %>% 
  select(-c(quarter, age, sample)) %>%
  filter(year %in% c(2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019),
         purpose == "Holiday") %>% 
  group_by(country) %>% 
  summarise(total_spend = sum(spend)) %>%
  slice_max(total_spend, n = 20) %>% 
  ggplot() +
  geom_col(aes(x = reorder(country, total_spend), y = total_spend,
               fill = country %in% c("USA")),
           show.legend = FALSE, colour = "white") +
  scale_fill_manual(values = c("FALSE" = "grey60",
                               "TRUE" = colour_scheme[1])) +
  coord_flip() +
  labs(
    x = "Country",
    y = "Total Expenditure (Million GBP)",
    title = "International Tourism: Total Expenditure (2002-2009)",
    subtitle = "Top 20 (Holiday Only)") +
  theme(plot.title = element_text(size = 15, face = "bold"),
        plot.subtitle = element_text(size = 10),
        panel.background = element_rect(fill = "white"),
        panel.grid = element_line(colour = "grey90", linetype = "blank"))

Expenditure per Visit by Country of Residence

Bar Plot

international_visits %>% 
  select(-c(quarter, age, sample)) %>%
  filter(year %in% c(2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019),
         purpose == "Holiday") %>% 
  group_by(country) %>% 
  summarise(total_spend = sum(spend),
            total_visits = sum(visits),
            spend_per_visit = total_spend / total_visits) %>%
  filter(total_visits > 500) %>%
  ggplot() +
  geom_col(aes(x = reorder(country, spend_per_visit), y = spend_per_visit,
               fill = country %in% c("Australia", "China", "Canada")),
           show.legend = FALSE, colour = "white") +
  scale_fill_manual(values = c("FALSE" = "grey60",
                               "TRUE" = colour_scheme[1])) +
  coord_flip() +
  labs(
    x = "Country",
    y = "Total Expenditure (Million GBP) / Total Visits (Thousands)",
    title = "International Tourism: Total Expenditure per Visit (2002-2009)",
    subtitle = "Countries w/ Over 500 Thousand Total Visits (Holiday Only)") +
  theme(plot.title = element_text(size = 15, face = "bold"),
        plot.subtitle = element_text(size = 10),
        panel.background = element_rect(fill = "white"),
        panel.grid = element_line(colour = "grey90", linetype = "blank"))

Scatter Plot

international_visits %>% 
  select(-c(quarter, age, sample)) %>%
  filter(year %in% c(2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019),
         purpose == "Holiday") %>% 
  group_by(country) %>% 
  summarise(total_spend = sum(spend),
            total_visits = sum(visits),
            spend_per_visit = total_spend / total_visits) %>%
  filter(total_visits > 500) %>% 
  ggplot(aes(x = total_visits, y = spend_per_visit)) +
  geom_point(aes(colour = country %in% c("Australia", "China", "Canada")), 
             show.legend = FALSE,
             size = 4) +
  scale_colour_manual(values = c("FALSE" = "grey60",
                                 "TRUE" = colour_scheme[3])) +
  ggrepel::geom_text_repel(aes(x = total_visits, y = spend_per_visit, label = country)) +
  labs(
    x = "Visits (Thousands)",
    y = "Total Expenditure (Million GBP) / Total Visits (Thousands)",
    title = "International Tourism: Visits vs. Expenditure Per Visit (2002-2009)",
    subtitle = "Countries w/ Over 500 Thousand Total Visits (Holiday Only)") +
  theme(plot.title = element_text(size = 15, face = "bold"),
        plot.subtitle = element_text(size = 10),
        panel.background = element_rect(fill = "white"),
        panel.grid = element_line(colour = "grey90", linetype = "dashed"))

Marketing Scotland To Canadian Based Tourists

\(H_0\): \(\mu_{\textrm{spend per visit(Q1+Q4)}} - \mu_{\textrm{spend per visit(Q2+Q3)}} = 0\) \(H_a\): \(\mu_{\textrm{spend per visit(Q1+Q4)}} - \mu_{\textrm{spend per visit(Q2+Q3)}} != 0\)

international_visits %>%
  filter(country == "Canada",
         purpose == "Holiday") %>% 
  mutate(exp_per_visit = spend / visits,
         spring_summer = if_else(quarter %in% c("Quarter 2", "Quarter 3"), "Yes", "No")) %>%  
  group_by(spring_summer) %>% 
  summarise(mean_visits = mean(visits),
            mean_exp = mean(spend),
            mean_exp_per_visit = mean(exp_per_visit))

canada_sample <- international_visits %>%
  filter(country == "Canada",
         purpose == "Holiday") %>% 
  mutate(exp_per_visit = spend / visits,
         spring_summer = if_else(quarter %in% c("Quarter 2", "Quarter 3"), "Yes", "No"))

observed_stat <- canada_sample %>% 
  specify(exp_per_visit ~ spring_summer) %>%
  calculate(stat = "diff in means", order = c("Yes", "No"))

observed_stat

Response: exp_per_visit (numeric)
Explanatory: spring_summer (factor)

null_distribution <- canada_sample %>% 
  specify(response = exp_per_visit, explanatory = spring_summer) %>%
  hypothesize(null = "independence") %>%
  generate(reps = 1000, type = "permute") %>% 
  calculate(stat = "diff in means", order = c("Yes", "No"))

null_distribution %>%
  visualise() +
  shade_p_value(obs_stat = observed_stat, direction = "both")

null_distribution %>%
  get_p_value(obs_stat = observed_stat, direction = "both")